This study addresses the challenges of extracting and analyzing biomedical information from vast amounts of unstructured text data using Large Language Models (LLMs). With the advent of big data and advancements in AI, traditional methods of manual data processing are no longer sufficient. Leveraging LLMs, such as the Qwen 115B model, specifically tailored for the biomedical domain, we propose a novel method that facilitates efficient knowledge extraction and integration. This method not only automates the extraction of valuable medical knowledge from unstructured texts but also structures this knowledge into a knowledge graph, aiding clinical decision-making and scientific research. Through specialized fine-tuning and advanced frameworks for information extraction, our approach demonstrates superior performance in handling biomedical text data. Comparative experiments and real-world applications validate the effectiveness of our method in tasks such as diagnostic assistance, drug discovery, and disease prediction, showcasing its potential impact on healthcare and biomedical sciences.